Towards Query-Agnostic RAG Evaluation via Query Coverage and Claim Verifiability
This episode discusses the paper 'Towards Query-Agnostic RAG Evaluation via Query Coverage and Claim Verifiability,' which introduces Q-CARE. This framework provides a reliable, automated method for scoring Retrieval-Augmented Generation (RAG) systems. It assesses whether an AI's answer covers all parts of a query and verifies if every claim is supported by the source documents, offering detailed results that align with human judgment.